It’s time to iterate on the product formerly known as the RSS reader. Twitter, Facebook, and Tumblr are going in a direction that emphasizes usability and ephemerality over durable value and utility. I want someone to do to the RSS reader what Apple has done to the iPhone. The iPhone is a phone — but it’s also a completely different paradigm.
For the longest time (since freshman year, early 2007), I’ve been dedicated to Google Reader. I briefly switched to Fever and switched back because the mobile interface wasn’t as functional. Given this whole new Google+ thing, I have less and less confidence of serious, necessary product innovation happening with Google Reader.
How I use Google Reader
Currently, Reeder, an excellent app for both Mac OS X and the iPhone, is my primary interface for Google Reader. It’s lovely, syncs well, and is amazingly integrated with other services. Click a button, choose a service, presto.
I’m subscribed to a total of 568 subscriptions, everything from friends to thought leaders to product blogs to aggregators. Surprisingly, not a lot of mainstream news organizations. My subscriptions are bundled into folders: A-List, B-List, B-High (pretty much Techmeme and Mediagazer), and then by topic (Media & Journalism, WordPress, etc.) or context (friends, Automattic, projects, etc.).
A fair number of people I know have switched away from using an RSS reader in favor of Twitter (example), under the auspices that “if the news is that important, it will find me.” I understand the value of serendipity — I also want to be deliberate about the information I consume. I zero my subscriptions on an almost daily basis.
What a Reader should be
A reader is for engaging with information; it’s a tool for consuming, managing, and using knowledge. In addition to presenting new information to consume, I also want it to pay attention to, infer insights from, and make accessible in an evergreen matter what I’ve already read. For me, this presents the pinnacle of personal information management — an intelligent tool that can reinforce what I already know and help guide me towards what I need to know.
Dave Winer has been thinking about this for much longer than I have — read him first. Basically, unread counts and the rigamarole associated with subscribing to a new feed are the two big deal breakers for RSS readers. In addition to fixing this, I think the new reading interface should:
- Present content based on type. Blog posts should look like blog posts. Quotes shouldn’t look like blog posts… they should look like quotes! Statuses should look like statuses, videos should behave like videos, etc.
- Keep track of what I’ve read and emphasize what’s new to me. Similarly, pay attention to links within pieces of content. If multiple pieces include the same link, nest those together so I can quickly skim the conversation once.
- Offer insights into my information diet. Help me better understand what I’m consuming, how often I’m consuming it, and how I can better improve my consumption patterns.
- Sync states across platforms. It bugs me to no end when I read a Twitter mention on the web and the read state doesn’t persist to the mobile app. If you’re sending email notifications too, the act of reading an email notification should put the web notification should be put in the read state.
- Help me budget my time. Prioritize content when I only have 15 minutes to spare, or I’m at work. I’ve seen some websites show “estimated reading time” — take that further.
- Make it easy to engage with the content. Whether it’s sharing, favoriting, or commenting, let me interact within the reading interface. Reeder does the sharing brilliantly; I pick which services I use and the action is only a tap away.
- Help me remember what I know. I partially use WordPress for this (e.g. useful tools I’ve come across or smart information from Clay Shirky). There’s a certain amount of friction with this approach (both good and bad). It would be awesome if I had a secondary system for quickly accessing information I’ve previously come across. A search engine for information I’ve consumed.
That’s what I’ve got. Let’s start iterating.
10 Comments
I’ve been looking for a better Google Reader as well. To me the core flaw of RSS readers is that they break the basic notion of surfing the web in two ways:
1. Stripping away the look and feel of web sites (making them less interesting and less diverse).
2. When you click on a link, you’re taken away from the reader, back into regular web surfing mode (browser tabs have partially solved this problem, but it’s still a discontinuous mode of jumping in and out of the reader).
I completely agree, although #1 is a difficult problem to solve because you need to overcome a herky-jerky experience reading multiple sites. NewsBlur is one attempt at a solution.
I was a big fan of Google reader but since I have Flipboard I’m only using this. I’m currently working on a side project utilizing arc90s readability stuff and bundling together all the links I receive in my feeds and twitter and other streams to rate the content I receive and present me by the stuff which seems interesting for me and also making sure that I will only receive each item once and not for every person who ever linked it.
I am still using Google Reader and can’t imagine anything else.
The current Android app with a swipe left to right to go item to item is a perfect mobile experience.
I also use Reeder on the iPad but the speed at which I can J-J-J-S-J my feeds on a MacBook Air is unprecedented. The next item on the iPad Reeder app is still super awkward and a swipe down is way too much effort when reading fast.
Being able to go back and search my Starred items is a life saver.
Here are two cool things I have done recently with Reader.
1) Make a “too busy to read” tag, take a break from Reader for a few days, any feed that has over 10+ items tag with “–too busy to read” this will highlight feeds you have that are just too noisy for their own good, somewhat like shuu.sh and anytime you want to cut them out you can mark all read for that tag.
2) Add some blogs about some topics you want to learn about, this will make your feed reading not only informative but also a learning experience, conversely remove blogs about topics that are boring you, you can always re-add them its not like you are throwing them in the trash.
Thanks for starting this conversation about Reader.
I used to be a net news wire user. But my unreads got out of control. Now I use flipboard, tweetdeck, hacker news, quora and BBC. I don’t care about the transport method (rss), I care about relevance to me, which is managed by the people and topics I follow.
The truth is I shouldnt even be trying to absorb as much as I currently do, it’s a waste of time. There is just too much data out there, even compared to my net news wire days. So I’ll take my news when I can rather than having a looming queue of read-it-laters (which incidentally I do by faving tweets).
As Thorsten has said… now I have Flipboard and it’s great! It aggregates sources and apply filters and actions depending of the account quite good. Of course, nothing to say about the wonderful interface but… I there’s a missing piece of the puzzle: As the content is remastered and ordered someway I do not know, am I sure it is ALL the content? or maybe the app decides to skip some content? An advance setting options for choosing the priority, or special tip to apply OPTIONALLY by the user should do the work: sentiment, trending topic, …
This is one point I was trying to make: a reading interface should allow you to be deliberate about your information consumption. It should give you more power and flexibility, not less.
Sounds like I might need to play around with Flipboard 🙂
I like your long form posts. You should do more of this. Inspires better dialogue.
Daniel: this is a great write up. I’m product cheerleader for RSS Hero, where we’re trying to reboot innovation in the feed reading space.
Google Reader, with it’s API that was only ever unofficial, wasn’t / isn’t something to drive innovation. And yet, building a central feed repository is hard. But we’re trying to do it anyway.
I agree that the “stress” around unread counts is not good. At the same time, SOME feeds I really do want to know the exact unread. I think this goes to the type of content: I may really want to know unread in a feed of commits to a project I’m monitoring, or unread counts in a feed of support requests (the list goes on).
Your thoughts here are a great starting point for what we’re thinking with RSS Hero. I’d love for you to try it and tell us what you think.
One of the things we haven’t implemented yet is folders / labels. In part, because that’s not how WE read, and in part because we’ve been looking for the potential of a different paradigm. Could you expand on WHY you bundle them into folders?